Multi-objective optimal energy-efficient retrofit determination using hybrid urban building energy model:Considering uncertainties between models  

作　　者：Linxi Luo Hailu Wei Ziqi Lin Jiyuan Wu Wei Wang Yongjun Sun 

机构地区：[1]School of Architecture,Southeast University,Sipailou 2,Xuanwu District,Nanjing,210096,China [2]Key Laboratory of Urban and Architectural Heritage Conservation(Southeast University),Ministry of Education,Nanjing,China [3]Department of Architecture and Civil Engineering,City University of Hong Kong,Tat Chee Avenue,Kowloon,Hong Kong,China

出　　处：《Building Simulation》2025年第1期183-206,共24页建筑模拟(英文)

基　　金：sponsored by National Natural Science Foundation of China(Grant Nos.52394224 and 52208011).

摘　　要：Typical energy-efficient retrofit studies based on urban building energy models face challenges in quickly obtaining appropriate retrofit solutions and often ignore the unexpected outcomes caused by inherent model uncertainty.To solve it,this study proposes a decision support framework that integrates a hybrid urban building energy model(UBEM)method,NSGA-II,and TOPSIS to obtain rapidly the optimal energy-efficient retrofit solutions that take into account model uncertainty.The study took the building groups in Sipailou campus as a case study and identified 76“stable solutions”and 149“active solutions”that minimize energy consumption,carbon emission,and life-cycle cost(LCC)over 30 years from 40,353,607 retrofit schemes.Key findings include that when considering model uncertainty,the quantities,types,and ranks of optimal retrofit solutions have changed.When the error of baseline UBEM validation is within±5%and considering uncertainty transmission from energy simulation to ANN model,the energy-saving potential of optimal retrofit schemes has expanded from[63.78,65.05]%to[60,68.75]%,carbon-saving potential has shifted from[63.69,64.09]%to[59.92,67.79]%,and the LCC has changed from[−40.68,14.59]×10^(6)to[−38.25,16.97]×10^(6)Yuan.This study provides decision makers with a scientific approach to consider the potential uncertainties and risks associated with optimal retrofit solutions.

关 键 词：urban building energy model building retrofit model uncertainty machine learning multi-objective optimization 

分 类 号：TU201.5[建筑科学—建筑设计及理论]